• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2017년도 추계학술대회
• /
• pp.66-69
• /
• 2017

본 논문에서는 깊은 합성 곱 신경망 (Deep Convolutional Neural Network) 를 이용해서 SAR (Synthetic Aperture Radar) 영상의 반전 잡음 (speckle noise) 성분을 제거하는 기법을 제안하고자 한다. Deep Convolutional Neural Network는 이미지의 데이터 특성에 적합한 딥 러닝 방법이고, 이는 SAR 위성영상의 반전 잡음 제거에 사용해도 효과적이다. 반전 잡음 필터 모델 추정을 위한 학습은 임의로 반전 잡음을 합성한 트레이닝 이미지들과 원본 트레이닝 이미지들을 이용한 회귀모델을 통해 진행된다. 학습을 통해 얻은 반전 잡음 필터는 기존 알고리즘에 비해 우수한 외곽선 보존 성능을 나타냄을 확인하였다.

• International journal of advanced smart convergence
• /
• 제5권3호
• /
• pp.1-7
• /
• 2016

In existing Convolutional Neural Network (CNNs) for object recognition task, there are only few efforts known to reduce the noises from the images. Both convolution and pooling layers perform the features extraction without considering the noises of the input image, treating all pixels equally important. In computer vision field, there has been a study to weight a pixel importance. Seam carving resizes an image by sacrificing the least important pixels, leaving only the most important ones. We propose a new way to combine seam carving approach with current existing CNN model for object recognition task. We attempt to remove the noises or the "unimportant" pixels in the image before doing convolution and pooling, in order to get better feature representatives. Our model shows promising result with CIFAR-10 dataset.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제16권3호
• /
• pp.208-215
• /
• 2016

The lithography hotspot detection process is crucial for semiconductor design development process. But, the lithography hotspot detection using optical simulation method takes much time and it slowdown the layout design development cycle. Though the geometry based approach is introduced as an alternative, it still revealed low detection performance and sophisticated framework. To solve this problem, we introduce a deep convolutional neural network based hotspot detection method. Our method made better results in ICCCAD 2012 dataset. To reach this score, we used lots of technical effort to improve the result in addition to just utilizing the nature of convolutional neural network. Inspection region reduction, data augmentation, DBSCAN clustering helped our work more stable and faster.

• 인터넷정보학회논문지
• /
• 제19권4호
• /
• pp.35-44
• /
• 2018

Monitoring through Synthesis Aperture Radar (SAR) is responsible for marine safety from floating icebergs. However, there are limits to distinguishing between icebergs and ships in SAR images. Convolutional Neural Network (CNN) is used to distinguish the iceberg from the ship. The goal of this paper is to increase the accuracy of identifying icebergs from SAR images. The metrics for performance evaluation uses the log loss. The two-layer CNN model proposed in research of C.Bentes et al.[1] is used as a benchmark model and compared with the four-layer CNN model using data augmentation. Finally, the performance of the final CNN model using the VGG-16 pre-trained model is compared with the previous model. This paper shows how to improve the benchmark model and propose the final CNN model.

• 한국멀티미디어학회논문지
• /
• 제19권9호
• /
• pp.1659-1668
• /
• 2016

Understanding 3D structure of scenes is of a great interest in various vision-related tasks. In this paper, we present a unified approach for estimating depth from a single monocular image. The key idea of our approach is to take advantages both of parametric learning and non-parametric sampling method. Using a parametric convolutional network, our approach learns the relation of various monocular cues, which make a coarse global prediction. We also leverage the local prediction to refine the global prediction. It is practically estimated in a non-parametric framework. The integration of local and global predictions is accomplished by concatenating the feature maps of the global prediction with those from local ones. Experimental results demonstrate that the proposed method outperforms state-of-the-art methods both qualitatively and quantitatively.

• ETRI Journal
• /
• 제43권4호
• /
• pp.684-693
• /
• 2021

This paper presents a low-cost two-stage approximate multiplier for bfloat16 (brain floating-point) data processing. For cost-efficient approximate multiplication, the first stage implements Mitchell's algorithm that performs the approximate multiplication using only two adders. The second stage adopts the exact multiplication to compensate for the error from the first stage by multiplying error terms and adding its truncated result to the final output. In our design, the low-cost multiplications in both stages can reduce hardware costs significantly and provide low relative errors by compensating for the error from the first stage. We apply our approximate multiplier to the convolutional neural network (CNN) inferences, which shows small accuracy drops with well-known pre-trained models for the ImageNet database. Therefore, our design allows low-cost CNN inference systems with high test accuracy.

• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.399-410
• /
• 2021

Aiming at the problem that complex external variables in natural scenes have a greater impact on facial expression recognition results, a facial expression recognition method based on two-stream convolutional neural network is proposed. The model introduces exponentially enhanced shared input weights before each level of convolution input, and uses soft attention mechanism modules on the space-time features of the combination of static and dynamic streams. This enables the network to autonomously find areas that are more relevant to the expression category and pay more attention to these areas. Through these means, the information of irrelevant interference areas is suppressed. In order to solve the problem of poor local robustness caused by lighting and expression changes, this paper also performs lighting preprocessing with the lighting preprocessing chain algorithm to eliminate most of the lighting effects. Experimental results on AFEW6.0 and Multi-PIE datasets show that the recognition rates of this method are 95.05% and 61.40%, respectively, which are better than other comparison methods.

• 방송공학회논문지
• /
• 제24권2호
• /
• pp.357-360
• /
• 2019

본 논문은 임베디드 GPU에서 실시간 동작하는 딥 컨볼루션 뉴럴 네트워크(CNN) 기반의 보행자 탐지 기법을 제안한다. 제안하는 기법에서는 먼저 영상 내 보행자 크기에 대한 통계적 분석을 통해서 최적의 컨볼루션 층의 개수를 결정한다. 또한, 본 논문에서는 다중 스케일 CNN 학습 기법을 적용하여 영상 내의 보행자 크기 변화에 강인한 탐지 기법을 개발한다. 컴퓨터 모의실험을 통해 제안하는 알고리즘이 임베디드 GPU에서 실시간 동작하면서도 기존의 기법과 비교하여 평균적으로 높은 정확도를 보임을 확인한다.

• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1387-1395
• /
• 2018

Core algorithm of deep learning Convolutional Neural Network(CNN) shows better performance than other machine learning algorithms. However, if there is not sufficient data, CNN can not achieve satisfactory performance even if the classifier is excellent. In this situation, it has been proven that the use of transfer learning can have a great effect. In this paper, we apply two transition learning methods(freezing, retraining) to three CNN models(ResNet-50, Inception-V3, DenseNet-121) and compare and analyze how the classification performance of CNN changes according to the methods. As a result of statistical significance test using various evaluation indicators, ResNet-50, Inception-V3, and DenseNet-121 differed by 1.18 times, 1.09 times, and 1.17 times, respectively. Based on this, we concluded that the retraining method may be more effective than the freezing method in case of transition learning in image classification problem.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.344-346
• /
• 2019

본 논문에서는 Ultra-Wideband(UWB) 시스템에서 합성곱 신경망을 이용한 거리 추정 기법을 제안한다. 합성곱 신경망을 이용한 딥러닝 모델을 학습하는데 사용하는 학습 데이터는 MATLAB 프로그램을 통해 생성하였으며, IEEE 802.15.4a 표준을 활용한다. 기존 거리 추정에 사용하는 문턱값 기반의 거리추정 기법과 성능 비교를 통해 제안하는 거리 추정 기법의 성능을 검증한다.